The purpose of this project is to explore the reasons for language- learning difficulties in children with specific language impairment. These children have recently been shown to have considerable difficulty learning a nonsense morpheme similar to real morphemes, but this difficulty is apparent only when the child is not allowed to imitate the adult model. Performance of these children in an imitation condition was similar to normal controls. In the current research, variations of this experimental procedure and different kinds of subjects will be used to search for the source of this imitation superiority. One task will replicate the original study. A second task will vary only in that the representational relation to be learned will be in the visual, rather than the auditory modality. The final task will be concept learning task with cognitive demands similar to the language learning tasks, but without the representational component. All three tasks will be presented in both a modeling and imitation format. There will be four groups of subjects: two groups of SLI children (receptive and expressive disorders, respectively), aged 5 to 6 years; a group of age- and IQ-matched normal children, and a group of younger normal children matched to the SLI children on language maturity (MLU) and IQ. By contrasting subjects' performances across these six tasks, it will be possible to determine whether the rule-learning difficulties of the SLI children represent a specific auditory- linguistic deficit, a linguistic deficit not tied to a perceptual modality, or a nonlinguistic-conceptual deficit. Individual case studies of selected subjects will be used to supplement the experimental findings. The results of the proposed studies should provide information concerning the underlying nature of specific language impairment that will prove useful in designing comprehensive intervention programs for this population.